Effects of Biologically Active Metabolites of Tamoxifen on the Proliferation Kinetics of MCF-7 Human Breast Cancer Cells in Vitro

[CANCER RESEARCH 43, 4618-4624, October 1983]

Roger R. Reddel, Leigh C. Murphy, and Robert L. Sutherland1

Ludwig Institute for Cancer Research (Sydney Branch), University of Sydney, Sydney, New South Wales 2006, Australia

ABSTRACT tumor tissues of patients treated with tamoxifen is DMT, with 4OHT being present at much lower concentrations (1, 7,13, 25). The effects of two major metabolites of tamoxifen, /V-de- Prompted no doubt in part by the early misidentification of the methyltamoxifen (DMT) and 4-hydroxytamoxifen (4OHT), on major metabolite as 4OHT, there has been considerable interest MCF-7 Å“il proliferation and cell cycle kinetic parameters were in the antiestrogenic and antitumor activity of 4OHT (2, 3, 6,15, compared with those of the parent compound. All three com 16, 33). In contrast, there have been very few studies on the pounds produced dose-dependent decreases in the rate of cell biological activity of DMT (6, 33). proliferation which were accompanied by decreases in the per In the present study, we have attempted to answer the follow centage of S- and G2-M-phase cells. 4OHT was 100- to 167-fold ing questions. Do the metabolites DMT and 40HT differ from more potent than both tamoxifen and DMT in producing these tamoxifen in their actions on the proliferation and cell cycle effects, and this was correlated with their relative binding affini kinetics of human breast cancer cells? What are the relative ties (RBAs) for the cytoplasmic estrogen receptor (ER) (17/3- potencies of these 3 compounds in their actions on such cells in estradiol = 100, 4OHT = 41, tamoxifen = DMT = 2). At doses vitro? What therefore is the likely contribution of each to the <2.5 Â¿IM,theseeffects were completely reversed by 170-estra- antitumor activity observed in vivo? Can the relative potencies diol, but the required 170-estradiol:antiestrogen concentration of these compounds in producing these effects be related to ratios differed, i.e., 1:10 to 1:1 for 4OHT compared with 1:1000 their RBAs for the ER or the more recently described high-affinity to 1:100 for tamoxifen and DMT. Although the concentrations of antiestrogen binding site (23, 30)? 170-estradiol required for reversal were related to affinity of the It has previously been shown in this laboratory that tamoxifen metabolite for ER, they were 5- to 20-fold lower than predicted has growth-inhibitory effects on the MCF-7 human breast cancer from the measured RBAs. When the rate of cell proliferation was cell line and that not all of these effects are reversed by simul measured over a range of concentrations of antiestrogen, in the taneous treatment with 17/3-estradiol (27, 28). These inhibitory presence or absence of 17/Ã®-estradiol,it was highly correlated effects were shown to be accompanied by accumulation of cells (r2 = 0.96) with the percentage of S-phase cells. In addition to in the GO-G! phase of the cell cycle (27, 28), due mainly to an these 17/3-estradiol-reversible events, all three compounds increase in the proportion of "slowly cycling" cells (28, 32). The caused 17j8-estradiol-irreversible cytotoxicity at higher concen MCF-7 cell line forms a particularly suitable experimental model trations (>7.5 U.MDMT and 4OHT, 10 MMtamoxifen). The order to compare the effects of tamoxifen and its metabolites since it of potency in producing this effect was DMT > 4OHT > tamox has previously been shown that metabolism of tamoxifen is ifen, which correlated with neither the RBAs for ER nor the RBAs negligible in these cells in vitro (3,14). for the high-affinity microsomal antiestrogen binding site. These data support the concept that estrogens and antiestro- MATERIALS AND METHODS gens compete for a common event which regulates the rate of cell proliferation probably by controlling the proportion of cells Cells. MCF-7 cells (26) in their 299th passage were supplied by Dr. entering S phase. Although it appears likely that ER is intimately Charles M. McGrath, Meyer L. Prentis Cancer Center, Detroit, Mich., involved in this regulatory process, 17ÃŸ-estradiol-irreversible and maintained as described previously (28). Stock cells were passaged weekly with an inoculation density of 3 x 106 cells/150-sq cm flask into mechanisms are also involved in antiestrogen action in vitro. 50 ml Roswell Park Memorial Institute Medium 1640 supplemented with 20 mw 4-(2-hydroxyethyl)-1-piperazineethanesulfonicacid buffer, 14 mM INTRODUCTION sodium bicarbonate, 5 mw L-glutamine, gentamicin (20 Â¿tg/ml),insulin (10 Mg/ml), and 10% fetal calf serum. Tamoxifen, a nonsteroidal antiestrogenic compound, has been Drug Treatment. Tamoxifen [frans-1-(4-/3-dimethylaminoethoxy- widely used in the treatment of human breast cancer on account phenylM ,2-diphenylbut-1-ene], N-demethyltamoxifen [frans-1-(4-,8-meth- of its efficacy and relatively low toxicity (22). Although this agent ylaminoethoxyphenyl)-1,2-diphenylbut-1-ene], and 4-hydroxytamoxifen has been in clinical use for over a decade, the mechanism of its [frans-1-(4-0-dimethylaminoethoxyphenyl)-1-(4-hydroxyphenyl)-2-phen- action remains incompletely understood (29). ylbut-1-ene] were supplied by I. C. I., Ltd., Pharmaceuticals Division, The original studies of tamoxifen metabolism identified 4OHT2 Macclesfield, Cheshire, United Kingdom. Stock solutions (10 mw) were prepared in analytical grade ethanol and stored at -20Â°. These drugs as the major metabolite (11, 12), but more recently it has been shown that the major metabolite accumulating in the plasma and were added to culture medium as described previously for tamoxifen (28). 17/3-Estradiol was acquired and handled as described previously (28). 1To whom requests for reprints should be addressed. Growth Curves. Exponentially growing cells (5 x 104) from Day 5 2 The abbreviations used are: 40HT, 4-hydroxytamoxifen; DMT, N-demethyltamoxifen; RBA, relative binding affinity; ER, estrogen receptor. stock cultures (28) were inoculated into 25-sq cm flasks (Coming, N. Y.) Received April 15, 1983; accepted July 13, 1983. in 5 ml medium containing 5% fetal calf serum. When cell numbers

4618 CANCER RESEARCH VOL. 43

X1Ã–5- 5X106-

K>6-

5XW5-

1 10=- 3" 5X104-

10 8

104-

103- 103-

30 96 144 96 144 HOURS HOURS

5X10

10' 30 100 500 2500 5.000 10000 HOURS CONCENTRATION - nM Chart 1. Effect of tamoxifen (A),DMT (B), and 40HT (C)on the growth of MCF-7 cells. After 30 hr of exponential growth, cells were treated with tamoxifen, DMT, or 4OHT at concentrations of 0 (â€¢).0.01(O), 0.1 (A), 0.5 (A), 2.5 (â€¢),5(D), 7.5 (Â»),or10 ^M (O). Treated medium was changed daily thereafter, and cells were harvested and counted at 96 and 144 hr. D, cell counts at 144 hr for tamoxifen (â€¢),DMT(A), and 4OHT (O).Points, mean of triplicate flasks from 2 experiments; bars, S.E. (n - 6). Inset, 2 to 10 nM4OHT compared with 1 pMtamoxifen at 120 hr; columns, mean of quadruplicate flasks; oars, S.E. reached approximately 1 x 10'yflask, the medium was replaced with supplied by I. C. I., Ltd., Pharmaceuticals Division, and frans-4-|3H|- fresh medium containing antiestrogen with or without 17/3-estradiol. The hydroxytamoxifen was prepared from this material by a modification of experimental medium was changed daily thereafter. Cells were harvested the high-performance liquid chromatography method of Foster ef al. (10) with 0.05% trypsin:0.02% EDTA (Flow Laboratories, Ryde, New South as follows. Isotope was applied to a Browntee RP-18 reverse-phase 10- Wales, Australia), and viable cells were counted under phase-contrast fim (22- x 0.7-cm) column and eluted in 75% methanol:25% water:0.1% on a hemocytometer or in a Model ZBI Coulter Counter (Coulter Elec diethylamine at 2.0 ml/min using a Waters Associates M45 solvent tronics, Hertfordshire, United Kingdom) after 2 and/or 4 doubling times delivery system. Column eluent was monitored by absorbance at 254 of the control population. nm and by radioactivity. The retention time of the c/s isomer was 24% Flow Cytometry. Harvested cells were stained for DMA flow cytometry longer than the frans isomer. The radioactive peak corresponding to as previously described (31) and analyzed on an ICP22 pulse cytometer /rans-4-hydroxytamoxifen was collected, dried under nitrogen, and redis- (Ortho Instruments, Westwood, Mass.). Estimates of the proportions of solved in ethanol. Aqueous solutions of fra/is-4-[3H]hydroxytamoxifen cells in Go-Gi, S, and G?-M phases of the cell cycle were computed from were prepared by drying a portion of the ethanolic solution under nitrogen the resulting DMA histograms (21). and redissolving in 10 mM Tris: 1.5 HIM EDTA buffer, pH 7.4 (TrisrEDTA Isotopes. [3H]Estradiol (100 Ci/mmol) was purchased from Amersham buffer) containing 0.2% bovine serum albumin. These solutions were International, Amersham, Buckinghamshire, United Kingdom, and [3H]- used within 24 hr of preparation. tamoxifen (71 Ci/mmol) from New England Nuclear, Boston, Mass. [3HJ- Determination of Binding Parameters. The cells used for binding 4OHT (42 Ci/mmol), a 1:1 mixture of the c/s and frans isomers, was studies were obtained by plating 2 x 106 logarithmically growing cells

OCTOBER 1983 4619

into 150-sq cm flasks in 50 ml medium and harvesting 6 to 7 days later nm ir Â§+ 1/1000 E2 m + 1/10E2 when cell numbers were 3 to 4 x 107 cells/flask. Cells were harvested 1-H/1E2!=:!!;::i~--_-, il +1/100E2 ^ with 1 rriM EDTA in Dulbecco's phosphate-buffered saline (140 ITIM 3X106-* NaCI:2.7 mM KCI:8.1 rtiM Na2HPO4:1.5 rriM KH2PO4, pH 7.3) and washed .; once with phosphate-buffered saline and once with Tris: EDTA buffer, -X.x![;i':'â€¢â€¢:;:â€¢'i':;;.;;.â€¢';â€¢::SE!5:.:::.:.â€¢:â€¢".â€¢:;':â€¢Au.||EÂ¡E;EEEE;;1 each wash being followed by centrifugation at 800 x g for 10 min. The cells were homogenized in Tris:EDTA buffer (3 x 107 cells/ml) with a 106-(/>Â¿ Teflorrglass homogenizer, and a cytosol was prepared by centrifugation at 135,000 xgfor 1 hr. Endogenous estrogens were removed from the cytosol by a 30 min 5X105-V)d0TO5-Â«;Yin4-n incubation with 0.1 volume 5% charcoal:0.5% dextran followed by a 10- min centrifugation at 1500 x g. RBAs. To determine the ability of various ligands to compete for ER, a constant concentration of [3H]estradiol (0.5 nM) was incubated for 16 hr at 4Â°with increasing concentrations of unlabeled ligand. Data were plotted as percentage of tracer bound versus log ligand concentration, and RBAs were calculated using the method of Korenman (18). Apparent Affinity Constants. Saturation analysis of binding to ER using increasing concentrations of (rans-4-[3H]hydroxytamoxifen (0.05 to 4 nM), [3H]tamoxifen (0.25 to 15 nM), or [3H]estradiol (0.05 to 2 nM) O 0.1 0.5 2.5 was performed by incubating 100 n\ cell extract with 50 /Â¿Iisotope and 50 n\ buffer or 50 n\ unlabeled ligand (4 MM)for 16 to 20 hr at 4Â°.Bound CONCENTRATION - and unbound ligand were separated by a 30-min incubation at 4Â°with 500 /il 0.5% charcoal:0.05% dextran followed by centrifugation at 1500 x g for 10 min at 4Â°.Data were analyzed by the method of Scatchard 3X10b-106-5X10-105-e following correction for nonspecific binding (5).

RESULTS

Effects of Tamoxifen Metabolites on Cell Proliferation. The effects of tamoxifen and the 2 metabolites, DMT and 4OHT, on the growth of MCF-7 cells in medium supplemented with 5% PCS are shown in Chart 1. Control cells grew exponentially with a doubling time of about 28 hr. Each of these 3 compounds caused a dose-dependent decrease in the rate of cellular prolif eration. The lowest concentration of tamoxifen and DMT found to produce a significant decrease in cell numbers after 114 hr of treatment was 100 nM, and these 2 compounds were approxi mately equipotent over the concentration range 100 nM to 5 JIM (Chart 1D). In contrast, 4OHT was greater than 50-fold more potent at the lower concentrations, with 10 nM 4OHT producing v\rf-L|'S|ii"?;E;:;Â£;ii1mii:i.i;fâ€¢-'iBâ€¢ii:EIâ€¢si 0.1 0.5 more inhibition than 500 nM tamoxifen or DMT (Chart 1, C and CONCENTRATION - D). The potencies of all 3 compounds in inhibiting cell growth tended to converge at concentrations of 2.5 to 5 UM, and at higher concentrations cytotoxicity was observed with DMT and 3X10-* 4OHT (7.5 to 10 MM),both causing a rapid decline in cell numbers which was significantly greater than the cytotoxicity produced by 10 MMtamoxifen (Chart 1). To further define the relative potencies of 40HT and tamoxifen, io6-(ASÃŒ 1 MMtamoxifen was compared with 4OHT over a concentration

=--IÃˆE:_E1s1|X^I'////////,sti;1[ii;i.'.'..:';:;â€¢;â€¢'.\Â¡.f\ range of 2 to 10 nM (Chart 1D, inset). This dose of tamoxifen ^|S inhibited cell proliferation to the same extent as 6 to 10 nM 4OHT, 5X105-C/i_J0105-C i.e., 4OHT was 100- to 167-fold more potent than tamoxifen, under these experimental conditions. i"11Ci;:.;.1i- Reversibility by Estradiol. In order to assess the ability of 17/3-estradiol to reverse the inhibitory effects of tamoxifen and its 2 metabolites on MCF-7 cell proliferation, cells were treated simultaneously with 17/3-estradiol and antiestrogen (0.1 to 5 MM) at relative concentrations ranging from 1:1000 to 1:1 (Chart 2). 1ilJ- Chart 2. Effect of tamoxifen (/(), DMT (B), and 40HT (C) with or without 17/3- v-liVO estradiol on MCF-7 cell growth. Cells were treated simultaneously with 17/3- Alt),t[*IiIâ€¢.-"v:â€¢:-;;!â€¢;â€¢:':i:â€¢'j"\ J1^|1sss1$1t 0.5 i-1N estradid and tamoxifen, DMT, or 4OHT at relative concentrations of 1:1000,1:100, 1:10, or 1:1. Treated medium was changed daily. Cells were harvested and counted CONCENTRATION - after 114 hr drug treatment. Columns, mean of quadruplicate flasks; bars, S.E.

4620 CANCER RESEARCH VOL. 43

Under these conditions, 170-estradiol alone produced either a 100 small but significant increase [16 Â±2% (S.E.); n = 36; p < 0.025] 90- in cell numbers over control when administered at 10~10- 2.5 x 80- 10~6 M or caused a small decrease (24 Â±6%; n = 6; p < 0.05) 70- at a dose of 5 x 10~6w. Ã ? 60- The effects of tamoxifen and DMT (0.1 to 2.5 JIM) were fully W 50" reversed by 1,000-fold lower concentrations of 17/Ã®-estradiol â€¢Â¿40- and, indeed, for 0.1 and 0.5 /Â¿Mtamoxifen and DMT, simultane 0 30- ous administration of 170-estradiol at relative concentrations of 20- 1:100 to 1:1 produced a small but significant stimulation of cell 10- proliferation (Chart 2, A and B) not exceeding that seen with the 0- same concentrations of 17/Ã®-estradiolalone. Almost complete 0.01 0.1 0.5 2.5 5 CONCENTRATION - reversal of the effect of 5 MMtamoxifen was produced by 100- and 10-fold lower concentrations of 17/3-estradiol (50 to 500 nw), 100 but no reversal was seen with equimolar 17/3-estradiol (Chart 90- 2A). Inhibition produced by 5 MMDMT was incompletely reversed 80- by 1000- to 10-fold lower concentrations of 17/3-estradiol (50 to 70- 500 nw), and the addition of equimolar 17/3-estradiol caused 60- further inhibition of cellular proliferation (Chart 20). 50- Â« In contrast, the inhibitory effect of 0.1 and 0.5 MM4OHT was j 40- fully reversed only by equimolar or 10-fold lower concentrations U 30- of 17/3-estradiol (Chart 2C). Complete reversal of the growth- 20- inhibitory effect of 2.5 UM 4OHT was not apparent at any 10- concentration of 17/3-estradiol tested; however, 0.25 MM 17/Ã®- 0- i Â»\ estradiol resulted in partial reversal to approximately 70% of 0 0.01 0.1 0.5 25 5 CONCENTRATION -i control cell numbers. In the presence of 5 MM4OHT, equimolar or 10-fold lower concentrations of 17/3-estradiol caused inhibition 100 additional to that produced by 5 MM4OHT alone (Chart 2C). 90- This latter effect was seen at all concentrations of antiestrogen 80- which induced a decline in cell numbers, i.e., no reversal was 70- apparent at any 17/3-estradiol concentration tested, and at some Â¿Ã§60- concentrations 17/3-estradiol augmented this cytotoxic effect of W 5Â°" antiestrogens (data not shown). J 40- Cell Cycle Parameters. Flow cytometric analysis of drug- 0 30- treated cells demonstrated that all 3 compounds caused a dose- 20- dependent reduction in the proportion of cells in S phase and, to 10- a lesser extent, in G2-M phase of the cell cycle, with a corre sponding increase in the proportion of cells in G0-Gi phases 0 0.01 0.1 0.5 (Chart 3). Over the concentration range shown (0.01 to 5 IIM), CONCENTRATION - Â¿iM 40HT was at least 50-fold more potent than both tamoxifen and Chart 3. Effects of tamoxifen (A), DMT (ÃŸ),and 4OHT (C) on the cell cycte DMT in causing these changes in cell cycle parameters, but parameters of MCF-7 cells. Cells were grown as described in Chart 1, harvested at 144 hr. and subjected to flow cytometry. The ONA histograms were analyzed tamoxifen and DMT were approximately equipotent. as described under 'Materials and Methods," and the percentage of cells in G, (x), In order to assess the estrogen reversibility of these effects, S (â€¢),andG;.-M (A) phases of the cell cycle were plotted. Points, mean of duplicate cells were treated simultaneously with 17/3-estradiol and anties flasks from 3 separate experiments (n = 6); oars, S.E. Where oars are not shown, the S.E. is smaller than the symbol. trogen at a relative concentration ratio of 1:10, and cell cycle kinetic parameters were calculated after approximately 4 dou blings of the control population. Under these conditions, 17/3- 4) appeared to parallel the changes in cell numbers seen in earlier estradiol alone (10~9 to 5 x 10~7M)caused a small but significant experiments (Chart 2). To elucidate the nature of this correlation, rise in the proportion of cells in S phase (5 Â±1%; n = 45; p < cell numbers and DNA histograms were obtained for cells treated 0.01). Complete reversal by 17/3-estradiol of the drug induced with antiestrogens (0.01 to 5 MM)in the presence or absence of effects on percentage of S-phase cells was seen with concentra a 10-fold lower concentration of 17/Ã®-estradiol(1to 500 nM). The tions up to 2.5 MMtamoxifen, 0.5 MM DMT, and 0.5 UM 4OHT, population-doubling time was calculated and hence the doubling with small increases over control being seen for tamoxifen (0.01 (proliferation) rate. Over the concentration range tested, the to 2.5 MM)plus 17/3-estradiol and for DMT (0.01 and 0.1 MM)with proliferation rate was found to correlate with the logarithm of 17/3-estradiol (Chart 4). Estradici only partially reversed the ef percentage of S-phase cells, with a coefficient of determination, fects seen with 5 MMtamoxifen, 2.5 and 5 MMDMT, and 2.5 MM r2 = 0.96 (Chart 5). 4OHT. The effect of 5 UM 4OHT on percentage of S-phase cells Binding of Tamoxifen Metabolites to ER. Direct saturation was unaffected by the presence of 500 nw 17/3-estradiol. analysis data for the binding of tritiated tamoxifen, trans-4- These patterns of reversal by 17/3-estradiol of the cell cycle hydroxytamoxifen, and 170-estradiol to MCF-7 cell cytosol kinetic changes induced by tamoxifen and its metabolites (Chart showed that each compound was bound to a single saturable

OCTOBER 1983 4621

metabolites (6) on the growth kinetics of human breast cancer i TAMOXIFEN 840HT IHDMT cells in vitro. In agreement with the findings of others, these data illustrate that tamoxifen is a potent inhibitor of cell proliferation in vitro (4,6,14,19), that many of its effects can be reversed by simultaneous administration of 17/3-estradiol (6, 19), and that 4OHT, which has enhanced affinity for ER (2, 6, 16, 33), is a more potent growth inhibitor than tamoxifen In vitro (6). More importantly, this study reveals that DMT and 4OHT have both estrogen-reversible and estrogen-irreversible components to their inhibition of cell proliferation in vitro, with the former being highly correlated with affinity for ER, that both compounds demonstrate cytotoxicity at doses >5 IÃ•Mbut with different potencies, and that both compounds induce changes in cell cycle kinetics similar to those seen with tamoxifen, i.e., an increase in the proportion of G0-Gi-phase cells at the expense of S- and G2- 0.01 0.1 0.5 2.5 5 CONCENTRATION -f>M

Chart 4. Effects of tamoxifen, DMT, and 40HT, with or without 17/3-estradiol, on the percentage of MCF-7 cells in S phase of the cell cycle. After 30 hr of exponential growth, cells were treated with tamoxifen, DMT, or 4OHT with or without a 10-fold tower dose of 17/3-estradiol. Treated medium was changed daily. Cells were harvested at 144 hr and subjected to flow cytometry as described under â€¢MaterialsandMethods.' Columns, percentage of cells in S phase of the cell cycle, mean of duplicate flasks from 3 experiments (n = 6); bars, S.E.

0.1 0.2 0.3 0.4 B- nM

100 40

Charts. Correlation between MCF-7 proliferation rate and the proportion of cells in S phase of the cell cycle following antiestrogen treatment. Cells were treated with tamoxifen, DMT, or 40HT (0.01 to 5 >M) with or without a 10-fold tower dose of 17/3-estradiol as described in Chart 4, and at 144 hr cells were harvested and counted, and the percentage of S-phase cells was calculated from DNA histograms obtained by flow cytometry. Points, mean of duplicate flasks from 3 experiments (n = 6) for cells treated with tamoxifen (â€¢),tamoxifen and 17/3-estradiol (A), DMT (x), DMT and 17/3-estradiol (O), 4OHT (O), or 4OHT and 170-estradiol (A); coeffi cient of determination, r" = 0.96. 60- binding component, presumably the ER (Chart 6A). The apparent equilibrium dissociation constant, K(/, was lowest for estradici, 0.10 nM, while frans-4-hydroxytamoxifen was bound with about 40 one-half this affinity, i.e., K

4622 CANCER RESEARCH VOL. 43

M-phase cells (27, 28). In the estrogen-reversible range where In addition, the antiproliferative effect was always fully or partially cytotoxicity is not evident, the degree of change in cell cycle reversed by 17|8-estradiol (Chart 2) while the cytotoxic effect parameters was correlated with affinity for ER, and in addition was either unaltered or augmented by the addition of 17/3- the rate of cell proliferation for all 3 antiestrogens over a 50-fold estradiol. concentration range, in the presence or absence of a 10-fold The ability of 17/3-estradiol to fully or partly reverse the anti lower dose of 17/3-estradiol, was highly correlated with the proliferative but not the cytotoxic effects of these compounds proportion of cells in S phase. This suggested that estrogen and raises the question of the involvement of ER in these effects. antiestrogen were competing for a common pathway that deter The relative concentrations of 17/3-estradiol required to reverse mined the rate of entry into S phase and thus the rate of cell the antiproliferative effects (Chart 2) are correlated with RBAs proliferation. Furthermore, DMT, the quantitatively most impor for ER (Chart 6). 4OHT required a higher relative 17/3-estradiol tant metabolite in human plasma and tumor tissue (7), was shown concentration than tamoxifen or DMT to effect maximal reversal. to be at least equipotent with tamoxifen in affecting cell prolifer [The exception was 5 /Â¿M4OHT,which was partially reversed by ation kinetics, which raises the possibility of a major role for this a 1000-fold lower concentration of 170-estradiol, although this metabolite in tumor responses to tamoxifen in vivo. Many of relative 17/3-estradiol concentration was ineffective at lower these points require further expansion and discussion. 4OHT concentrations (Chart 2C); we are unable to explain this The present experimental design with daily medium changes apparent anomaly.] However, the relative concentrations of 17/3- differs from our previous study (28) where medium remained estradiol required for complete reversal were 5- to 20-fold lower unchanged and drug unreplenished over a 6-day period. Inter than would be predicted from the measured RBAs of 17/3- estingly, the sensitivity to tamoxifen was very similar in both estradiol and antiestrogen for ER. The explanation for this is studies, adding support to the conclusion that tamoxifen is not unclear but could include the following, (a) Occupation of all ER rapidly metabolized or inactivated in this system (14). sites may not be essential for maximal MCF-7 proliferation. Data reported here for the affinity of 4OHT and tamoxifen for Indeed, Lippman ef al. (19) reported maximal induction of MCF- the cytoplasmic ER from MCF-7 cells are in good agreement 7 cell proliferation at concentrations of 17/3-estradiol sufficient to with those reported by others (6, 8). However, the estimates of saturate only about 10 to 15% of ER sites, (o) Incomplete affinity for 170-estradiol are somewhat higher than reported by solubility of the antiestrogens in culture medium (6) may account others (4, 8), leading us to the conclusion that in MCF-7 cells for some of the decrease in expected activity relative to 17/3- 4OHT is bound somewhat less tightly than 17/3-estradiol. This is estradiol. (c) Binding studies done in cell-free systems at lowered in agreement with the competitive binding data presented in temperatures may not reflect the situation in whole cells where Chart 66 and in Ref. 6. The observation that tamoxifen and DMT there could be differential permeability of cells to 17/3-estradiol have almost identical affinities for ER agrees with other published and antiestrogens or compartmentalization of antiestrogens studies (6, 33). within the cell, e.g., by binding to the high-affinity microsomal It has previously been observed in this laboratory that tamox antiestrogen binding site (23, 30). ifen (27,28) and clomiphene (24) have cytotoxic effects on MCF- The involvement of ER in the cytotoxic effect, however, seems 7 cells as well as dose-dependent inhibitory effects on cell unlikely in view of its 17/3-estradiol Â¡(reversibilityand the lack of proliferation. This is also true for DMT and 4OHT since cell death correlation between relative potency and affinity for ER. There clearly plays a major role in the rapid decrease in cell numbers has, however, been some interest recently in the possibility that seen at concentrations >5 pM (Chart 10). It is not yet clear tamoxifen may act through its high-affinity binding site since a whether the cytotoxicity observed in vitro has a correlate in vivo; tamoxifen-resistant MCF-7 line with similar ER concentration to however, tamoxifen has been shown to cause cell death in the wild type had markedly decreased levels of antiestrogen uterine glandular epithelial cells in rats (20). In addition, the binding site (17). In addition, a study from this laboratory showed reported mean intratumor concentration of tamoxifen was 25 a possible relationship between the cytotoxic properties of some ng/mg protein (7); assuming a tumor protein content of 5 to clomiphene analogues and their affinity for this antiestrogen 10%, this is equivalent to 3.4 to 6.7 /Â¿Mtamoxifen. The mean binding site (24). The present data do not, however, support intratumor concentration of DMT was more than twice this. The such a relationship since the relative potency of these com sum of the intratumor concentrations of tamoxifen and DMT in pounds in producing cytotoxicity did not correlate with their vivo was thus in the concentration range at which tamoxifen is affinity for the antiestrogen binding site (tamoxifen > 4OHT > cytotoxic to MCF-7 cells in vitro. The mean plasma concentra DMT3). tions, however, of tamoxifen and DMT were reported as 0.8 and Assessment of the likely relative contributions of tamoxifen 1.3 UM, respectively (7). Which of these concentrations in vivo and its metabolites to its antitumor effects in vivo depends on most closely resembles the situation in vitro awaits direct mea which, if any, of the effects in vitro correlate with tumor regres surement of antiestrogen concentration within MCF-7 cells under sion in vivo. If the noncytotoxic antiproliferative effect is relevant, the present assay conditions. then consideration of both the relative levels of 4OHT, tamoxifen, There are several lines of evidence indicating that these anti- and DMT (7) in human tumors (0.02:1:1.5) and their relative proliferative and cytotoxic effects have separate mechanisms. potencies (150:1:1) would indicate that all 3 compounds may be 40HT is at least 100-fold more potent than both tamoxifen and important, with relative contributions of 3:1:1.5, respectively. If DMT in inducing the antiproliferative effect (Chart 1), while the the cytotoxic effect is more relevant, however, DMT is likely to order of potency in producing the cytotoxic effect is DMT > 4OHT > tamoxifen (Chart 1D). These data, in agreement with a 3C. K. W. Watts, L. C. Murphy, and R. L. Sutherland, Microsomal binding sites for nonsteroidal antiestrogens in MCF-7 human mammary carcinoma cells: dem previous study on clomiphene analogues (24), demonstrate that onstration of high affinity and narrow specificity for baste ether derivatives of the structural specificity for producing these 2 effects is different. tnphenylethylene, manuscript in preparation.

OCTOBER 1983 4623

Downloaded from cancerres.aacrjournals.org on October 2, 2021. © 1983 American Association for Cancer Research. R. R. Reddel et al. be the major antitumor agent, with a lesser contribution from in patients with breast cancer: comparison of traditional and loading dose schedules. Cancer Treat. Rep., 64: 765-773, 1980. tamoxifen. 10. Foster, A. B., Griggs, L. J., Jarman, M., Van Maanen,J. M. S., and Schulten, An understanding of the relative contributions of these metab H. R. Metabolism of tamoxifen by rat liver microsomes: formation of the A/ olites to the action of tamoxifen is important for interpretation of oxide, a new metabolite. Biochem. Pharmacol.,29:1977-1979,1980. 11. Fromson, J. M., Pearson, S., and Bramah, S. The metabolism of tamoxifen tamoxifen pharmacokinetics in humans. During chronic admin (I.C.I. 46,474). Part I: in laboratory animals.Xenobiotica, 3: 693-709, 1973. istration, tamoxifen reaches steady-state levels after 4 weeks, 12. Fromson, J. M., Pearson, S., and Bramah, S. The metabolism of tamoxifen (I.C.I. 46,474). Part II: in female patients. Xenobiotica,3: 711-714,1973. in contrast to the 8 weeks required for DMT (25). This means 13. Golander, Y., and Sternson, L. A. Paired-ion Chromatographieanalysis of that the maximal combined effects of tamoxifen and its metab tamoxifen and two major metabolites in plasma. J. Chromatogr., 787:41-49, olites may not be reached until much later than previously 1980. assumed, and this gives added weight to the proposal for 14. Horwitz, K. B., Koseki, Y., and McGuire, W. L. Estrogen control of progester one receptor in human breast cancer: role of estradiol and antiestrogen. tamoxifen "loading" schedules (9, 34). Endocrinology, 703: 1742-1751, 1978. In summary, these data show that the 2 metabolites, DMT 15. Jordan, V. C., and Allen, K. E. Evaluationof the antitumour activity of the nonsteroidal antiestrogen monohydroxytamoxifen in the DMBA-inducedrat mam and 40HT, have similar actions to tamoxifen on MCF-7 cells in mary carcinoma model. Eur. J. Cancer, 76: 239-251, 1980. vitro in that they have 170-estradiol-reversible and 17/3-estradiol- 16. Jordan, V. C., Collins, M. M., Rowsby, L., and Prestwich, G. A monohydroxy- irreversible components to their growth-inhibitory activity that lated metabolite of tamoxifen with potent antioestrogenic activity. J. Endocri- nol., 75:305-316,1977. are associated with accumulation of cells in the G0-Gi phase of 17. Jozan, S., Elalamy, H., and Bayard, F. Ã‰tudedumÃ©canismed'actiond'un the cell cycle. Regardless of which of these components may be antioestrogÃ¨nedu groupe triphÃ©nylÃ©thylÃ¨nesurla croissance de la lignÃ©e cellulaire de cancer du sein humain MCF 7 en culture. C. R. Hebd. SÃ©ances relevant to the antitumor action of tamoxifen in vivo, considera Acad. Sci., 292: 767-770, 1981. tion of the reported intratumor concentrations of the metabolites 18. Korenman, S. G. Relation between estrogen inhibitory activity and binding to (7) together with their relative potency in producing growth cytosol of rabbit and human uterus. Endocrinology,87: 1119-1123,1970. 19. LJppman,M. E., BolÃ¡n,G., and Huff, K. The effects of estrogens and anties inhibition in vitro leads to the conclusion that at least one of trogens on hormone-responsive human breast cancer in long-term tissue these metabolites is likely to make a major contribution to the culture. Cancer Res., 36: 4595-4601, 1976. clinical antitumor action of tamoxifen. Since the potency of these 20. Martin, L. Effects of antioestrogenson cell proliferationin the rodent reproduc tive tract. In: R. L. Sutherland and V. C. Jordan (eds.), NonsteroidalAntioes compounds in producing the 17/S-estradiol-reversible growth in trogens, pp. 143-163. New York: Academic Press, Inc., 1981. hibition is correlated with affinity for ER, it is likely that an ER- 21. Milthorpe, B. K. FMFPAKI: a program package for routine analysis of single parameter flow microfluorimetric data on a low cost mini-computer. Comput. mediated mechanism is a major regulator of the distribution of cells between the previously described "slowly cycling" and Biomed. Res., 73: 417-429,1980. 22. Mouridsen, H., Palshof, T., Patterson, J., and Battersby, L. Tamoxifen in "rapidly cycling" pools (28, 32) and thus the rate of cell prolifer advanced breast cancer. Cancer Treat. Rev., 5:131-141, 1978. ation. The 17/Ã®-estradiol-irreversible, cytotoxic component, how 23. Murphy, L. C., and Sutherland,R. L. Modificationsin the aminoetherside chain of clomipheneinfluenceaffinity for a specific antiestrogen binding site in MCF ever, seems unlikely to be ER mediated since no such correlation 7 cell cytosol. Biochem. Biophys. Res. Commun., 700: 1353-1360,1981. could be found between the potency of these compounds in 24. Murphy, L. C., and Sutherland, R. L. Antitumor activity of clomipheneanalogs in vitro: relationship to affinity for the estrogen receptor and another high producing this effect and their affinities for ER. affinity antiestrogen-binding site. J. Clin. Endocrinol. Metab., 57: 373-379, 1983. 25. Patterson, J. S., Adam, H. K., Kemp, J. V., and Settatree, R. S. Serum ACKNOWLEDGMENTS concentrations of tamoxifen and major metabolite during long-term Nolvadex therapy, correlated with clinical response. In: J. T. Mouridsen and T. Palshof The authors thank I.C.I., Ltd., PharmaceuticalsDivision,for supply of tamoxifen, (eds.), Breast Cancerâ€”Experimentaland Clinical Aspects, pp. 89-92. New DMT, 4OHT, and [3H]40HT. The interest and critical comments of Professor M. H. York: Pergamon Press, 1980. N. Tattersali are gratefully acknowledged. 26. Soule, H. D., Vazquez, J., Long, A., Albert, S., and Brennan, M. A human cell line from a pleural effusion derived from a breast carcinoma. J. Nati. Cancer Inst., 57:1409-1413,1973. REFERENCES 27. Sutherland, R. L., Green, M. D., Hall, R. E., Reddel, R. R., and Taylor, I. W. Tamoxifen induces accumulation of MCF 7 human mammary carcinoma cells 1. Adam, J. K., Douglas, E. J., and Kemp, J. V. The metabolism of tamoxifen in in the GO/G!phase of the cell cycle. Eur. J. Cancer Clin. Oncol., 79:615-621, humans. Biochem. Pharmacol.,28: 145-147, 1979. 1983. 2. Borgna, J. L, and Rochefort, H. High-affinitybinding to the estrogen receptor 28. Sutherland, R. L., Hall, R. E., and Taylor, I. W. Cell proliferation kinetics of of pH^-hydroxytamoxifen, an active antiestrogen metabolite. Mol. Cell. En- MCF-7 human mammary carcinoma cells in culture and effects of tamoxifen docrinol., 20: 71-85, 1980. on exponentially growing and plateau-phase cells. Cancer Res., 43: 3998- 3. Borgna, J. L., and Rochefort, H. Hydroxylated metabolites of tamoxifen are 4006,1983. formed in vivo and bound to estrogen receptor in target tissues. J. Biol. Chem., 29. Sutherland, R. L., and Murphy, L. C. Mechanisms of oestrogen antagonism 256:859-868,1981. by nonsteroidalantioestrogens. Mol. Cell. Endocrinol.,25: 5-23,1982. 4. Butler, W. B., Kelsey, W. H., and Goran, N. Effects of serum and insulin on 30. Sutherland, R. L., Murphy, L. C., Foo, M. S., Green, M. D., Whybourne,A. M., the sensitivity of the human breast cancer cell line MCF-7 to estrogens and and Krozowski, Z. S. High affinity antioestrogen binding sites distinct from the antiestrogens. Cancer Res., 41: 82-88, 1981. oestrogen receptor. Nature (Lond.),288: 273-275,1980. 5. Chamness, G. C., and McGuire, W. L. Scatchard plots: common errors in 31. Taylor, I. W. A rapid single step staining technique for DNA analysis by flow correction and interpretation. Steroids, 26: 538-542, 1976. microfluorimetry.J. Histochem.Cytochem., 28:1021-1024,1980. 6. Coezy, E., Borgna, J. L., and Rochefort, H. Tamoxifen and metabolites in 32. Taylor, I. W., Hodson, P. J., Green, M. D., and Sutherland, R. L. Effects of MCF-7 cells: correlation between binding to estrogen receptor and inhibition tamoxifen on cell cycle progression of synchronous MCF-7 human mammary of cell growth. Cancer Res., 42: 317-323, 1982. carcinoma cells. Cancer Res., 43: 4007-4010,1983. 7. Daniel, C. P., Gaskell, S. J., Bishop, H., Campbell, C., and Nicholson, R. I. 33. Wakeling, A. E., and Slater, S. R. Estrogen-receptor binding and biologic Determinationof tamoxifen and biologicallyactive metabolites in human breast activity of tamoxifen and its metabolites. Cancer Treat. Rep., 64: 741-744, tumours and plasma. Eur. J. Cancer, 17: 1183-1 189, 1981. 1980. 8. Eckhert, R. L., and Katzenellenbogen, B. S. Physical properties of estrogen 34. Wilkinson, P. M., Ribiero, G. G., Adam, H. K., Kemp, J. V., and Patterson, J. receptor complexes in MCF-7 human breast cancer cells. Differences with S. Tamoxifen (Nolvadex) therapyâ€”rationalefor loading dose followed by anti-estrogen and estrogen. J. Biol. Chem., 257: 8840-8846, 1982. maintenancedose for patients with metastatic breast cancer. Cancer Chem- 9. Fabian, C., Stemson, L., and Barnett, M. Clinical pharmacology of tamoxifen other. Pharmacol., 70: 33-35, 1982.

4624 CANCER RESEARCH VOL. 43

Downloaded from cancerres.aacrjournals.org on October 2, 2021. © 1983 American Association for Cancer Research. Effects of Biologically Active Metabolites of Tamoxifen on the Proliferation Kinetics of MCF-7 Human Breast Cancer Cells in Vitro

Roger R. Reddel, Leigh C. Murphy and Robert L. Sutherland

Cancer Res 1983;43:4618-4624.

Updated version Access the most recent version of this article at: http://cancerres.aacrjournals.org/content/43/10/4618

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://cancerres.aacrjournals.org/content/43/10/4618. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.